Notable and welcome sustained trends in the polling on immigration, as highlighted in the Red Box email today:
It’s a good example too of how the canny way to read opinion poll findings is to think not only about what they currently say but also about how likely they are to change or to be wrong (hence how the general election result this year didn’t catch me out that much).
It’s like being on a bus journey and seeing no congestion ahead of the bus. That’s good, but if you want to know if the bus will arrive on time you also need to factor in what may or may not happen to change that clear road and how big an impact it’s likely to have.
A classic example (and one where this error is made much less often) is the poll ratings of party leaders. You need to factor in the don’t knows to have a sense of how the current net figures may change.